基于改进YOLOv5和ResNet50的女装袖型识别方法OACSTPCD
A method for identifying women's sleeves based on improved YOLOv5 and ResNet50
针对女装袖型分类繁多、特征识别困难、检测效果不理想等问题,根据不同女装袖型的关联信息,结合注意力机制改进的YOLOv5 目标检测网络和ResNet50 残差网络,提出了一种女装袖子造型的自动识别方法.首先,从电商平台收集服装样本图像,按照长短大类和形态小类标记对女装袖型进行归类,建立了包含 3600 张图像的袖型数据集;其次,结合注意力机制改进的YOLOv5 目标检测网络和ResNet50 残差网络,设计了女装袖型识别方法;最后,在袖型数据集上开…查看全部>>
In response to the problems of numerous classifications of women's clothing sleeve shapes,difficult feature recognition,and unsatisfactory detection results,a deep learning method based on improved YOLOv5 and ResNet50 was used to achieve automatic recognition of women's clothing sleeve shapes on the basis of fully utilizing the correlation information between different women's clothing sleeve shapes. Firstly,two methods of sleeve type classification were com…查看全部>>
曹涵颖;妥吉英
重庆第二师范学院美术学院,重庆 400065重庆理工大学车辆工程学院,重庆 400054
轻工业
女装袖型深度学习YOLOv5注意力机制ResNet50
women's sleeve shapedeep learningYOLOv5attention mechanismResNet50
《现代纺织技术》 2024 (1)
45-53,9
重庆市教科委"十四五"规划项目(2021-JZ-030)重庆市教委科学技术研究计划项目(KJQN202201621)重庆市自然科学基金项目(cstc2020jcyj-msxmX0331)重庆第二师范学院校级科研项目(KY202127C)
评论